Model List API

On this page

List models

http

GET /open/v1/models

Returns all models currently available to call. Models, parameters and prices change with platform configuration, so read this endpoint in your program to determine the available parameters instead of hard-coding them.

curl https://api.zroimg.com/open/v1/models \
  -H "Authorization: Bearer zk-xxxxxxxxxxxxxxxxxxxxxxxx"

Response

json

{
  "object": "list",
  "data": [
    {
      "id": "gpt-image-2",
      "object": "model",
      "name": "GPT Image 2",
      "type": "image",
      "modes": ["text_to_image", "image_to_image", "inpaint"],
      "ratios": ["1:1", "16:9", "9:16", "4:3", "3:4"],
      "qualities": ["low", "medium", "high", "auto"],
      "counts": [1, 2, 3, 4],
      "maxReferenceImages": 1,
      "defaults": { "mode": "text_to_image", "ratio": "1:1", "quality": "low", "count": 1 },
      "pricing": {
        "unit": "image",
        "items": [
          { "mode": "text_to_image", "quality": "low", "points": 10 },
          { "mode": "text_to_image", "quality": "medium", "points": 15 }
        ]
      },
      "description": "Default image model"
    },
    {
      "id": "YOUR_VIDEO_MODEL_ID",
      "object": "model",
      "name": "Example Video",
      "type": "video",
      "modes": ["text_to_video", "image_to_video"],
      "ratios": ["16:9", "9:16"],
      "resolutions": ["720p"],
      "durations": [4, 6, 8],
      "maxReferenceImages": 9,
      "defaults": { "mode": "text_to_video", "ratio": "16:9", "resolution": "720p", "duration": 6 },
      "pricing": {
        "unit": "request",
        "items": [
          { "mode": "text_to_video", "resolution": "720p", "points": 40, "unit": "request" },
          { "mode": "image_to_video", "resolution": "720p", "points": 40, "unit": "request", "reference_image_points": 0 }
        ]
      }
    }
  ]
}

Prices in the example only illustrate the format; actual values are those returned by the endpoint.

Model object

Field	Type	Description
`id`	string	Model ID (same as on the web app), passed as the `model` parameter when submitting a task
`object`	string	Always `model`
`name`	string	Display name
`type`	string	`image` or `video`
`modes`	string[]	Supported generation modes; see the table below
`ratios`	string[]	Supported aspect ratios, e.g. `16:9`
`qualities`	string[]	Image models only: supported quality levels
`counts`	number[]	Image models only: number of images that can be generated per request
`resolutions`	string[]	Video models only: supported resolutions
`durations`	number[]	Video models only: supported durations (seconds)
`maxReferenceImages`	number	Maximum number of reference images; 0 means reference images are not supported
`defaults`	object	Default values used for parameters omitted on submission
`pricing`	object	Billing method and prices; see below
`description`	string	Optional model description

Generation modes

Mode	Type	Description
`text_to_image`	Image	Generate images from a prompt only
`image_to_image`	Image	Generate from 1 reference image
`inpaint`	Image	Inpainting: 1 source image plus a `mask_url` mask
`text_to_video`	Video	Generate videos from a prompt only
`image_to_video`	Video	Generate videos from reference images

Image qualities low, medium and high correspond to roughly 1K, 2K and 4K output; auto is billed as medium.

Pricing fields

pricing.unit determines how credits are calculated:

image: credits per image × number of images n.
second: credits per second × video duration duration.
request: a fixed number of credits per video.

pricing.items lists the unit price points by mode and quality (images) or by mode and resolution (videos). Image-to-video entries may include reference_image_points, the extra credits charged per reference image.

The credits actually charged are given by cost_points returned from the submit endpoint, under the same rules as the web app; see Credits and billing.